Metabolism of oxygen, while necessary for aerobic life, results in toxic by-products known as free radicals. Generation of these toxins is implicated in many diseases including cancer, AIDS, diabetes, arthritis, aging, stroke, and atherosclerosis. The body protects against these reactive oxygen species by producing enzymes known as glutathione peroxidases. While a deficiency of these enzymes is linked to disease development, their expression and regulation is not understood. Two glutathione peroxidases, gastro-intestinal (GPX2) and phospholipid (GPX4), differ from other GPXs in both tissue expression and substrate specificity. GPX2 protects against carcinogenic dietary lipid peroxidation products and GPX4 protects against oxidative mitochondrial-mediated apoptosis. Detailed studies exist on the biochemical function of GPX2 & GPX4, but little is known of regulatory mechanisms. Recent evidence indicates both GPX2 & GPX4 are regulated transcriptionally by oxidative-responding cis-elements. It has been repeatedly demonstrated that elevated levels of a specific GPX provide an effective cellular defense. Thus, understanding the normal mechanisms for stimulating endogenous GPX-based cellular defenses can yield a logical process for developing therapeutic regimens. To this goal, the aims of this proposal are designed to elucidate molecular mechanisms controlling stimuli-dependent GPX2 & GPX4 expression. We will identify cis- and trans-elements directing mitochondrial GPX4 production, which is critical in times of oxidative stress to block apoptosis. Oxidative-responsive elements (OREs) in the 5'-flanking regions of both GPX2 & GPX4, and their respective binding proteins, will be characterized and identified. Studies will determine if common peroxidase negative regulatory elements also function in GPX2 & GPX4, and cis-elements responsible for the unusual tissue distribution of GPX2 & GPX4 will be identified. An example is provided of how data obtained from this work can be developed into a potent tool for high throughput robotic screening of large combinatorial or natural product libraries, to identify potent pharmacological moieties.